4 ms·
I’ve tried gnucash a few times but it was just too much work for not enough rewards. I’ve contemplated just making my own scrape to database script because all
by prepend 5y ago
I’ve tried gnucash a few times but it was just too much work for not enough rewards. I’ve contemplated just making my own scrape to database script because all I really want is a ledger I can query.
Moneydance has been decent enough and I think I’ve paid $100-150 in 10+ years for an initial version and then an upgraded one.
- mindslight 5y agoThe download functionality of Moneydance and GNUCash is via "OFX Direct Connect", which is an older non-HTTPS protocol to download OFX data from a bank. But as web culture took over with things like Plaid, banks have become more wary about serving up a bespoke protocol that (I believe) authenticates in plain text, or at the very least doesn't support snake oil 2FA and the like. For example, Capital One 360 used to serve it up, but they turned it off in the past several years. Meanwhile, all banks seem to have the functionality to download .ofx/.qfx files through their web interface. I believe this is what Plaid et al retrieve. Of course using Plaid is horrible because not only are you giving out your credentials, but you're also giving away your private financial data to a surveillance company. But you can personally download the OFX files from the web interface for use with Moneydance (etc), and given the way OFX works you don't even need to be careful about overlapping date ranges (eg download the last year of transactions every month). Of course that becomes a bit tedious and doesn't give Moneydance (or whatever) access to your recent transactions (it would be very convenient to `tail /var/log/bank` rather than having to login manually). So I'm wondering if anyone has written a libre package that aims to do the web scraping approach for various popular banks that don't support Direct Connect.
- prepend 5y agoOFX supports SSL/TLS as the protocol. Doesn’t directly support 2fa, but I don’t care as it’s read only downloads of data. I still manually download csv or qif files, but that’s much less convenient than automatic imports. I like the idea of an open standard and wish banks would throw them away instead of improving them. OFX was issued in the 90s so the fact it runs at all is at least something.